So today I had a little spare time in the studio and decided to compare the RMS levels of some songs. If you don’t know what RMS is it means “Root Mean Square” and is useful for seeing how loud a piece of audio is. Personally, I am not for or against the loudness wars, rather I’d like to think some music just sounds better louder and some music sounds better softer. So you can say I am for whatever sounds good. However, I had to rub my eyes a few times looking at the RMS of some of my favorite songs.
One of them is A$AP Rocky’s PMW.
I love the bass in this song and the vocals are very clear. The song is also loud as hell! With a peak RMS of -4dbs this song would put a lot of heavy metal to shame. How can this song be this loud without turning into a pool of mush? Simplicity. It only contains about 4 sounds: vocals, 808 bass, drums and a high synth. That’s it. That’s the magic. The less sounds your song has the louder you can make it. Of course, loudness isn’t the goal. Making your song sound the best is the goal and should always be the goal. As I stated before some songs sound better louder and some sound better softer. Some of my favorite rock songs such as The Killer’s Sam’s Town have a peak RMS of -7.5dbs (which in 2006, when it was released, would be considered quite loud).
What RMS level should your songs be at before you send them to mastering?
Each mastering studio has different requirements that are based on personal preferences and on the equipment they’re using. I would say a good level pre-mastering for a mix would be around -14dbs. Keep in mind this isn’t a hard rule, if it sounds better at -12dbs than leave it at that. However, don’t send a song in at -3db and expect your mastering engineer to take you seriously. What are they going to do with that? You are basically shoving a pool of mush into a box to be squeezed into an even smaller box. Does that even sound right?
Peak RMS vs Average RMS
Another thing to keep in mind is the dynamic range of your song. A lot of heavy metal songs are unlistenable because their dynamic range is so small. For example, they’ll have a peak RMS of -2db and an average RMS of -4db. That’s a very small dynamic range, just – 2db, which basically means the song is all loud. Some of the best sounding songs have much bigger dynamic ranges say from about – 7db to -10db and it is because the loud parts and the soft parts contrast well with each other. Remember our ear only knows what is loud when we hear something soft. It is just like color. If you think the walls in your house are white, hold a white piece of paper up to it, doesn’t look so white anymore?
House songs have used dynamic range to ad infinitum by having these long drawn out parts with no bass or kick and that eventually lead to the cliché build up and then the boom. The reason it is so popular and works so well is because that dynamic range between the quiet part and loud parts is what builds emotion. Remember music is all about stirring emotion in the listener. Having a large dynamic range is one tool that can build anticipation and emotion.